Exo Labs
@exolabs
Followers
11,906
Following
3
Media
5
Statuses
90
Explore trending content on Musk Viewer
Build Wearing Denim
• 229310 Tweets
#RBSJackRusselXBuild
• 228371 Tweets
#KAMIGATA_BOYZ
• 107707 Tweets
BINI BACK ON GFEST
• 50839 Tweets
イタリア
• 46024 Tweets
オールカマー
• 43892 Tweets
神戸新聞杯
• 43010 Tweets
SmoothE x PondPhuwin
• 37707 Tweets
予約予定
• 34173 Tweets
ジャック
• 33935 Tweets
あんスタ
• 31963 Tweets
アニエラ
• 31200 Tweets
Vol.5
• 25669 Tweets
To the Edge
• 25290 Tweets
#Coldplay
• 23913 Tweets
メイショウタバル
• 23645 Tweets
#BookMyShow
• 22232 Tweets
ぱしゃっつ
• 19205 Tweets
レーベンスティール
• 18774 Tweets
殺傷事件
• 16321 Tweets
ラジオ体操
• 13787 Tweets
P.A.shots
• 13080 Tweets
#bbclaurak
• 10643 Tweets
マジック1
• 10568 Tweets
ツイステ
• 10430 Tweets
Pinned Tweet
The future of AI is open source and decentralized
"Exo's use of Llama 405B and consumer-grade devices to run inference at scale on the edge shows that the future of AI is open source and decentralized." -
@mo_baioumy
43
218
1K
5
30
215
Exo Labs Retweeted
Running Qwen 2.5 Math 72B distributed across 2 MacBooks.
Uses
@exolabs_
with the MLX backend.
5
12
102
Exo Labs Retweeted
The future of AI is open source and decentralized.
Amen.
22
58
552
Exo Labs Retweeted
yeah, this is genuinely good news for hardware to consumer providers
0
3
20
Exo Labs Retweeted
This figure from the o1 research report makes the case for a setup like this much stronger.
We can scale LLM capability with more inference time compute.
Search is sync-lite: you don't need ultra low latency / high bandwidth interconnect between your
@__tinygrad__
tinyboxes.
4
4
42
Exo Labs Retweeted
Think twice about trading in that old MacBook
8
7
164
Exo Labs Retweeted
Yes, exo will support Apple TV's.
21
17
369
Exo Labs Retweeted
On the power efficiency of Apple M chips.
@exolabs_
with the MLX backend can get ~100% GPU utilisation out of each MacBook Pro M3 Max. The GPU has 128GB of memory (bandwidth ~400GB/s) and consumes ~40 Watts of power with ~30 TFLOPs of fp16 performance.
MacBook Pro M3 Max GPU:
-
25
114
821
Exo Labs Retweeted
is there a name for this? when someone is shaking from using your product?
4
1
28
Exo Labs Retweeted
More progress in the
@exolabs_
pytorch interface
- Tested with Qwen2 Instruct, TinyLlama and Llama3 8b and was able to produce tokens
- Currently not producing good output and gibberish. Looking at my top_p, top_k and temp along with my logit_sample method
- Testing on
2
6
30
LLM on a plane.
they should just host the llms on planes. that way we won't need to connect to the internet
39
5
260
3
4
65
Exo Labs Retweeted
Managed to run Llama 3.1 405b on my flight.
Uses
@exolabs_
to distribute the AI model across 2 MacBooks.
GPT-4o-level model running offline, so I can play games, ask questions and use a coding assistant in the air.
77
67
1K
Exo Labs Retweeted
Got 2 MacBooks with me running distributed Llama-3.1-405b on
@exolabs_
ready for my flight.
Portable intelligence in the air, works fully offline with Thunderbolt 4 interconnect.
79
144
2K